Re: A conlang idea rolling around in my head

Gary Shannon wrote:
> --- "H. S. Teoh" <hsteoh@...> wrote:
>
>>>any pictograph could be mentally converted into a
>>>five-digit number literally at a glance, and this
>>>numerical key was the cannonical order for the glyph
>>>dictionary.
>>
>>That's neat.
>
> I call it the "LOTEP" number of a glyph. I count
> L = the number of complete Loops or enclosed areas
> (e.g., "P" and "#" have 1, "8" and "B" have two each).
> O = the number of Odd junctions (e.g. "T" has an odd
> number of lines entering the junction at the top).
> T = the number of Terminal, or endpoints, (e.g. "W"
> and "H" have 2 each while "Y" and "E" have 3
> endpoints).
> E = the number of even-parity junctions (e.g., "X" has
> one central junction with an even number of lines
> leading into it).
> P = the number of seperate Pieces (e.g. lower case "i"
> has two seperate, unattached pieces.)
>
> Odd and Even junctions are counted only for junctions
> with more than 2 lines entering them.  A junction with
> 2 lines in considered to be a continuous single line
> that turns a corner, and is not counted as a junction.
>
> "LOTEP" obviously stands for Loop, Odds, Terminals,
> Evens, Pieces.
>
> "H" has LOTEP = 02401
> "A" has LOTEP = 12201
> "W" has LOTEP = 00201
> "X" has LOTEP = 00411
> "%" has LOTEP = 20203
> If I superimpose the letters "O" and "X" that glyph
> has a LOTEP = 40451
> The Merceds Benz logo has LOTEP = 33001
> and the Volkwagen logo has LOTEP = 64031
>
> Many glyphs, especially simple ones, can share LOTEP
> numbers (e.g. "F", "Y" and "T" are all 01301) and so
> the numbers are used not to uniquely pinpoint a single
> glyph in the dictionary, but to get you to the right
> page where you may have to search through 6 or 8
> glyphs to find the one you are looking up.
That's an interesting system. More abstract than the radicals that kanji
dictionaries use.

Reply